A disease outbreak prediction system that adapts itself according to region and tries to predict the number of disease cases and the contribution of factors afftecting the same. The project introduces a new machine learning algorithm that can predict immediate future value of a variable with very high accuracy using a very limited amount of data. The algorithm doesn't requrie any parameter adjustment or data pre-processing, since it itself identifies which factors are relevant and what are their weightages.This project also gives a graphical analysis of disease count and affect of factors with respect to time.